1. Field of the Invention
The present invention relates to an image processing apparatus which processes an array of n original mount areas (n.gtoreq.2) of a subscanning direction each including a plurality of originals which are arranged in a line in a main scanning direction.
2. Description of the Background Art
There is a need for scanning an array of a number of originals and for storing resultant image signals (which express image data) in a storage medium such as a disk for original by original. One of conventional methods to meet this need is as follows.
Each original is scanned over its image area, all image processing steps necessary for each original are performed, and resulting image signals are stored as files. FIG. 7 shows an example of this. In FIG. 7, a first original is scanned and its image is read during a period from a time t.sub.s0 to a time t.sub.s1. Resultant image signals are stored in a memory and thereafter processed under an image processing condition which is set for this original. The processed image signals of the first original are stored in a disk as a file 1 during a period from the time t.sub.s1 to a time t.sub.F1. At the time tsl, an image reading head is moved to a second original. The second original is scanned and its image is read during a period from the time t.sub.s1 to a time t.sub.s2. In a similar manner, during a period from the time t.sub.s2 to a time t.sub.f2, processed image signals of the second original are stored in a disk as a file 2, while concurrently moving the head to a third original to start reading and processing of the third original. Such processes are performed in series until processing of the last original is complete. Scanning in the main scanning direction may be performed by a line scanning element such as a CCD installed in the head or a surface scanning element installed in the head. Scanning in the subscanning direction may be realized by moving the head in the subscanning direction. Scanning takes about 100 seconds, for example, whereas filing takes drastically shorter 24 seconds.
The conventional method above is not satisfactory since it is necessary to move the head for each original and a scanning time necessary for every original cannot be eliminated. It takes a long time to complete filing of image signals of all originals, which directly counters a recent need for high-speed mass processing.
One approach to deal with this is to define a scanning area in advance and to perform image processing while switching image processing conditions for the respective originals on a real time basis during scanning. See FIGS. 8A and 8B for details. In FIG. 8A, two originals A and B are placed next to each other. Assume that the originals A and B are being read by a head 60 along a main scanning line SL. Here, a switch 61 of FIG. 8B is switched under control every time the head 60 sweeps over intersections of the main scanning line SL and the originals A and B. In other words, at an intersection P1, the switch 61 is connected to an image processing circuit 62 (in which an image processing condition for the original A is set). While the head 60 moves from the intersection P1 to an intersection P2, the image processing circuit 62 processes image signals which are read by the head 60 and processed image signals are stored in a disk 65. At an intersection P3, the switch 61 is switched to an image processing circuit 63 (in which an image processing condition for the original B is set) so that the image processing circuit 63 processes image signals which are obtained between the intersection P3 and an intersection P4. The switch 61 is switched under the control of a CPU 64 which operates in accordance with signals outputted by the head 60 such as a position detect signal.
It is true that the approach above makes it unnecessary to move the head for each original, thereby reducing the total processing time. However, demanding a high main-processing speed and a switching circuit which can operate at an accordingly high speed and requiring a plurality of image processing circuits as described above, this approach to be realized without expensive hardware. To a further disadvantage, high-speed processing may not be possible at all because of slower storing of signals in the disk during filing. Thus, this approach is not practical and is not acceptable in any regard. It is therefore important to offer an improved technique which solves all of these problems at the same time.